Utilizing Prior Information on Tracking Moving Objects in Infrared Image Sequences

This paper introduces a suitable method for detecting and tracking moving objects throughout open areas such as country borders or sea coast. We utilized our pre-proposed segmentation algorithm [27] using the standard fuzzy c-means (FCM). The proposed method tends to increase robustness against noise and speed up the convergence of the foreground detection process. The FCM objective function is modified by incorporating a priori information related to the background. We used the spatial distribution information of the background without any moving objects call Atlas. This method aims to regularize the clusters produced by the FCM algorithm thus boosting its performance under noisy and unexpected data acquisition conditions. The detected moving objects are labeling and finding its centroid. The tracking process is achieved by calculating the difference between the locations of the object centroid in the present frame and the previous one.